The goal of this application is to develop imageable tumor-targeting bacteria that can cure tumors without progressive infection of the host. Previous experiments employed anaerobic microorganisms for cancer therapy but used a markedly different approach. Naturally nonvirulent or attenuated microbes served primarily as delivery vehicles for therapeutic genes. Target specificity appeared largely due to the anaerobic requirements met principally in necrotic tumor areas. The resulting tumor killing was at best limited. Therefore, more effective targeting is necessary, especially in the viable tumor tissue. Toward this goal, we have recently developed whole-body imaging systems that enable the visualization of green fluorescent protein (GFP)- and red fluorescent protein (RFP)-expressing tumors and bacteria (Lancet Oncology 3, 546-556, 2002; Proc. Natl. Acad. Sci. USA 98, 9814-9818, 2001). With the help of the imaging technology, we have developed a unique tumor-targeting Salmonella typhimurium strain (Proc. Natl. Acad. Sci USA 102, 755-760, 2005). This strain is an auxotrophic but fully virulent S. typhimurium variant termed A1. The A1 auxotrophic strain selectively grows in and destroys viable as well as necrotic malignant tissue but has little effect on normal tissue. The A1 bacteria eventually disappears from normal tissue even in immunodeficient nude mice. This remarkable selectivity apparently reflects the imposed nutritional requirements that are apparently met only in the cancer cell milieu. We rapidly selected this bacteria using assays of fluorescent tumor-bearing live mice. S. typhimurium A1 is a double amino acid auxotroph that requires Leu and Arg. Selecting therapeutic bacteria in live animal models would be prohibitively expensive, time-consuming, and extremely tedious, save for the real-time imaging afforded by the whole- body imaging systems to be used in this application. The specific aims are as follows: (1) Select additional multiple amino-acid auxotrophs of S. typhimurium to enhance tumor-killing selectivity and reduce possible host virulence in mouse tumor models; (2) Establish dependence of particular auxotrophic strains of S. typhimurium on tumor type, including patient-tumor mouse models; and (3) Determine possible significance of host immunological status by comparing antitumor efficacy of selected S. typhimurium auxotrophs in nude-mouse and immunocompetent-mouse tumor models. In the Phase II grant, the effective antitumor bacterial strains will be further developed using larger animals for eventual clinical application. Determination of possible synergy of tumor targeting S. typhimurium auxotrophic strains and chemotherapeutic agents as well as radiology will also be tested in the Phase II application. In addition, Salmonella amino acid tumor-targeting auxotroph strains will be modified to secrete cytokines to enhance the bacterial killing of the tumor with stimulation of host antitumor immune reactions in Phase II. [unreadable] [unreadable] [unreadable]